Abstract Data assimilation (DA) is a promising approach to improve meteorological and PM2.5 forecasts, but to what extent and by what process the DA of meteorological fields helps improve PM2.5 forecast still call for more discussion. By utilizing WRFDA and WRF-Chem models, we have assimilated AMSU-A, MHS radiances and conventional observations, and studied the influences of the meteorological DA on meteorological and PM2.5 forecasts over Central China through a series of experiments. The results show that multi-source meteorological DA helps improve temperature and relative humidity forecasts in the lower atmosphere, and the improved meteorological fields further improve PM2.5 forecast with a reduction of bias and RMSE by 7.4% and 4.1% over the study area, especially during PM2.5 episode. This study also helps understand how DA improve the PM2.5 forecasts over Central China.

中文翻译：

---

同化多源观测对华中地区气象及PM2.5预报的影响

摘要 数据同化（DA）是改善气象和 PM2.5 预报的一种很有前景的方法，但气象场的数据同化（DA）在何种程度上、通过什么过程有助于改善 PM2.5 预报仍有待进一步讨论。利用WRFDA和WRF-Chem模型，我们同化了AMSU-A、MHS辐射和常规观测，通过一系列实验研究了气象DA对华中地区气象和PM2.5预报的影响。结果表明，多源气象DA有助于改善低层大气温度和相对湿度预报，改进后的气象场进一步改善了PM2.5预报，研究区偏差和RMSE分别降低了7.4%和4.1%，特别是在 PM2.5 发作期间。这项研究还有助于了解 DA 如何改善 PM2.5。